Scopus İndeksli Yayınlar Koleksiyonu / Scopus Indexed Publications Collection

Permanent URI for this collectionhttps://hdl.handle.net/11147/7148

Browse

Filters

2010 - 201922020 - 20251

Settings

Access Right: Open AccessSubject: search.filters.subject.Electroencephalography

Search Results

Now showing 1 - 3 of 3
  • Article
    Vision-Language Model Approach for Few-Shot Learning of Attention Deficit Hyperactivity Disorder Using EEG Connectivity-Based Featured Images
    (IOP Publishing Ltd, 2025) Catal, Mehmet Sergen; Gumus, Abdurrahman; Karabiber Cura, Ozlem; Aydin, Ocan; Zubeyir Unlu, Mehmet
    Traditional medical diagnosis approaches have predominantly relied on single-modality analysis, limiting clinicians to interpreting isolated data streams such as images or time series. The integration of vision language models (VLMs) into neurophysiological analysis represents a paradigm shift toward multimodal diagnostic frameworks, enabling clinicians to interact with diagnosis models through diverse modalities including text, audio, visual inputs, etc. This multimodal interaction capability extends beyond conventional label-based classification, offering clinicians flexibility in diagnostic reasoning and decision-making processes. Building on this foundation, this study explores the application of VLMs to electroencephalography (EEG)-based attention deficit hyperactivity disorder (ADHD) classification, addressing a gap in neurophysiological diagnostics. The proposed framework applies VLM-based few-shot ADHD classification by converting raw EEG data into EEG connectivity-based featured images compatible with contrastive language-image pre-training's (CLIP) image encoder. The adaptor-based CLIP approach (Tip-Adapter and Tip-Adapter-F) for few-shot learning improves CLIP's zero-shot classification performance, achieving 78.73% accuracy with 1-shot and 98.30% accuracy with 128-shot using the RN50x16 backbone. Experiments investigate prompt engineering effects, backbone architectures of CLIP, patient-based classification, and combinations of EEG connectivity features. Comparative analysis is performed with two datasets to evaluate the approach between different data sources. Through the adaptation of pre-trained VLMs to neurophysiological data, this technique demonstrates the potential for multimodal diagnostic frameworks that enable flexible clinician-model interactions beyond conventional label-based classification systems. The approach achieves effective ADHD classification with minimal training data while establishing foundations for applying VLMs in clinical neuroscience, where diverse modality interactions through text, visual, and audio inputs can enhance diagnostic workflows. The code is publicly available on GitHub to facilitate further research in the field: https://github.com/miralab-ai/vlm-few-shot-eeg.
  • Article
    Citation - WoS: 24
    Citation - Scopus: 24
    Mice Doubly-Deficient in Lysosomal Hexosaminidase a and Neuraminidase 4 Show Epileptic Crises and Rapid Neuronal Loss
    (Public Library of Science, 2010) Seyrantepe, Volkan; Lema, Pablo; Caqueret, Aurore; Dridi, Larbi; Hadj, Samar Bel; Carpentier, Stephane; Boucher, Francine; Levade, Thierry; Carmant, Lionel; Gravel, Roy A.; Hamel, Edith; Vachon, Pascal; Di Cristo, Graziella; Michaud, Jacques L.; Morales, Carlos R.; Pshezhetsky, Alexey V.
    Tay-Sachs disease is a severe lysosomal disorder caused by mutations in the HexA gene coding for the a-subunit of lysosomal β-hexosaminidase A, which converts GM2 to GM3 ganglioside. Hexa-/- mice, depleted of b-hexosaminidase A, remain asymptomatic to 1 year of age, because they catabolise GM2 ganglioside via a lysosomal sialidase into glycolipid GA2, which is further processed by β-hexosaminidase B to lactosyl-ceramide, thereby bypassing the β-hexosaminidase A defect. Since this bypass is not effective in humans, infantile Tay-Sachs disease is fatal in the first years of life. Previously, we identified a novel ganglioside metabolizing sialidase, Neu4, abundantly expressed in mouse brain neurons. Now we demonstrate that mice with targeted disruption of both Neu4 and Hexa genes (Neu4-/-;Hexa-/-) show epileptic seizures with 40% penetrance correlating with polyspike discharges on the cortical electrodes of the electroencephalogram. Single knockout Hexa-/- or Neu4-/- siblings do not show such symptoms. Further, double-knockout but not single-knockout mice have multiple degenerating neurons in the cortex and hippocampus and multiple layers of cortical neurons accumulating GM2 ganglioside. Together, our data suggest that the Neu4 block exacerbates the disease in Hexa-/- mice, indicating that Neu4 is a modifier gene in the mouse model of Tay-Sachs disease, reducing the disease severity through the metabolic bypass. However, while disease severity in the double mutant is increased, it is not profound suggesting that Neu4 is not the only sialidase contributing to the metabolic bypass in Hexa-/- mice.
  • Article
    Citation - WoS: 15
    Citation - Scopus: 17
    Quasi-Supervised Learning for Biomedical Data Analysis
    (Elsevier Ltd., 2010) Karaçalı, Bilge
    We present a novel formulation for pattern recognition in biomedical data. We adopt a binary recognition scenario where a control dataset contains samples of one class only, while a mixed dataset contains an unlabeled collection of samples from both classes. The mixed dataset samples that belong to the second class are identified by estimating posterior probabilities of samples for being in the control or the mixed datasets. Experiments on synthetic data established a better detection performance against possible alternatives. The fitness of the method in biomedical data analysis was further demonstrated on real multi-color flow cytometry and multi-channel electroencephalography data. © 2010 Elsevier Ltd. All rights reserved.